{ "best_metric": 0.3143588602542877, "best_model_checkpoint": "./checkpoint-th/checkpoint-1500", "epoch": 0.2763894327106894, "eval_steps": 500, "global_step": 1500, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.09, "learning_rate": 1.963147226828819e-05, "loss": 4.6049, "step": 500 }, { "epoch": 0.09, "eval_bleu": 24.7133, "eval_gen_len": 33.9437, "eval_loss": 1.1478698253631592, "eval_runtime": 214.6895, "eval_samples_per_second": 4.714, "eval_steps_per_second": 1.178, "step": 500 }, { "epoch": 0.18, "learning_rate": 1.9262944536576377e-05, "loss": 0.3574, "step": 1000 }, { "epoch": 0.18, "eval_bleu": 25.1845, "eval_gen_len": 34.0662, "eval_loss": 0.31802815198898315, "eval_runtime": 214.1787, "eval_samples_per_second": 4.725, "eval_steps_per_second": 1.181, "step": 1000 }, { "epoch": 0.28, "learning_rate": 1.8894416804864568e-05, "loss": 0.1925, "step": 1500 }, { "epoch": 0.28, "eval_bleu": 25.5018, "eval_gen_len": 33.7075, "eval_loss": 0.3143588602542877, "eval_runtime": 211.869, "eval_samples_per_second": 4.777, "eval_steps_per_second": 1.194, "step": 1500 } ], "logging_steps": 500, "max_steps": 27135, "num_train_epochs": 5, "save_steps": 500, "total_flos": 1.04021020901376e+17, "trial_name": null, "trial_params": null }